Canadian Journal of Plant Science. Management of organic hairy vetch (Vicia villosa) cover crops in the establishment year

Transcription

1 Management of organic hairy vetch (Vicia villosa) cover crops in the establishment year Journal: Manuscript ID CJPS R1 Manuscript Type: Short Communication Date Submitted by the Author: 25-Aug-2016 Complete List of Authors: Bamford, Keith; University of Manitoba Entz, Martin; University of Manitoba, Plant Science Keywords: plant date, seed source, weed suppression

2 Page 1 of 12 Management of organic hairy vetch (Vicia villosa) cover crops in the establishment year Keith C. Bamford and Martin H. Entz 1 Department of Plant Science, University of Manitoba, Winnipeg, MB, R3T 2N2 Corresponding Author: M. H. Entz Keywords: Planting date; seed source; weed suppression Field studies showed that Hairy vetch (Vicia villosa) of northern origin produced more biomass than cultivars developed in milder climates. Hairy vetch biomass was greatest and weed biomass was lowest when vetch was planted in spring vs mid-summer. Approximately 50 plants m -2 maximized hairy vetch biomass production and weed suppression. Hairy vetch (Vicia villosa) is a plant of interest in organic production (USDA 2011) since it fixes high levels of N and provides weed suppression by forming surface mulch. The mulch provided by hairy vetch has facilitated organic no-till grain crop production (Halde et al. 2015). Hairy vetch was first described in Manitoba in the late 1800s at the Experimental Farm for Manitoba (Brandon Research Farm) where the plant was described to have a very spreading habit and would have been difficult to harvest (Bedford 1900). Spring planted hairy vetch results in large amounts of biomass (Halde et al. 2015) making it an excellent full-season legume cover crop; full season cover crops are typically used in Canadian prairie organic production (Entz et al. 2001), though hairy vetch use is still limited. 1

3 Page 2 of 12 Spring planted vetch in the prairie region withstands blade rolling in the year of establishment (Halde et al. 2015), which allows selective control of annual weeds that establish alongside the vetch. Hairy vetch then winterkills, resulting in dead mulch for no-till seeding the following spring (Halde et al. 2015). The prairie system of using a full-season legume cover crop differs from longer season regions, such as parts of eastern Canada. In the US Mid-Atlantic region, for example, hairy vetch is often planted in late summer after crop harvest (USDA, 2011; Maul et al. 2011). Under these conditions, hairy vetch survives the winter, and blade rolling has been used to kill vetch plants before organic crop (eg., corn Zea mays) planting the following spring (Mirsky et al. 2009). No previous studies on fall planted hairy vetch have been conducted in the prairie region, and so the first objective of this study was to determine growth and overwinter survival when planting hairy vetch at different times of the year. Previous research on the role of hairy vetch in organic no-till in Canada (Halde et al. 2015) was conducted with one northern origin hairy vetch genotype, yet genetic variation in hairy vetch exists (Maul et al. 2011). Therefore, a second objective was to compare different hairy vetch genotypes, in particular how vetches of northern origin compare with genotypes developed for milder climates (USDA, 2011). The final objective regards hairy vetch seeding rates because at presently recommended rates (approximately 25 kg seed ha -1 ), hairy vetch is among the most expensive legume cover crop options. Mirsky et al. (2012) tested different ratios of hairy vetch and triticale (Triticale ( Triticosecale)) in fall-planted system, however to our knowledge, no previous study has considered seeding rate with full-season hairy vetch systems. 2

4 Page 3 of 12 Field experiments were conducted for two years at Carman, Manitoba. The Chernozemic soil was from the Hochfeld series (Mills and Haluschak 1993) and vetch experiments were preceded by barley (2011) or oat (2012) silage crops. Experiment 1 consisted of 5 hairy vetch genotypes seeded at 3 different times of year (early spring - May 13 and 16, 2012 and 2013, respectively; mid-summer July 17 and 12, 2012 and 2013 respectively; and early autumn September 6 and August 30, 2012 and 2013, respectively). The experiment was arranged in a split plot with seeding date as the mainplot and genotype as the subplot. Four replicates were used. Two genotypes of northern origin were tested. The first was produced in central Minnesota (sourced from Albert Lea Seeds, Albert Lea, Minnesota; referred to as MN ). Seed of MN was produced at the University of Manitoba farm and is referred to as MB. Other genotypes were hairy vetches produced in Oregon (sourced from Homestead Organics, Berwick, Ontario; referred to as Oregon ) and Washington State (sourced from Pickseed, Winnipeg, Manitoba, referred to as Washington ). Two cultivars developed at USDA in Maryland (Purple prosperity and Purple bounty) were also included in the study. Hairy vetch genotypes of northern origin may have originated from the cultivar Madison, the only Vicia villosa ssp. villosa accession entry in the U.S. National Plant Germplasm System (2016). Madison has greater winter survival and more trichomes on the leaves and stems (Nicholas Wiering, University of Minnesota, pers. comm.). In the present study, MN and MB did indeed show more hairiness (more trichomes) than all other lines (K. Bamford, field observations). All plots were seeded with a seed drill (Fabro Industries, Swift Current, Saskatchewan) at a depth of 1.5 cm and 15 cm row spacing. Plot size dimensions were 1.8 m (12 rows) by 10 m. Seeds were inoculated with a granular pea (Pisum sativa) inoculant (EMD Crop BioScience Inc.). Plant population density was measured in 1 m length of row randomly selected within 3

5 Page 4 of 12 each plot 28 days after planting. An estimation of flowering was conducted visually with 1= no flowers and 5 = 100% of plants having flowers. Aboveground vetch and weed biomass were measured from 2, 0.25 m -2 samples per plot harvested at intervals (or just once) during the growing season. Weeds were separated from crop biomass, and both were dried at 80 C for 48 hours before weight measurement. In the first year of experiment one, aboveground biomass and plant density was measured for surviving plants in the early autumn seeding date treatment the following spring. Experiment two compared four different planting densities. The factorial experiment included 4 seeding rates (30, 60, 90 and 120 live seeds m -2 ), 2 vetch genotypes (Purple prosperity and Oregon) and was replicated 4 times. Crops were seeded on May 23 in both years. Plant population density was measured approximately 4 weeks after seeding using the procedure described for experiment one. Total vetch and weed biomass were measured in mid-august in both years and again in late fall using the procedure outlined for experiment one. Hairy vetch establishment was generally highest with earlier planting (Table 1). Lower plant establishment with later planting was attributed to hotter and dryer soil conditions in midsummer. No genotype effect was observed in A genotype x seeding date interaction in 2013 was due to Purple bounty and Oregon having lower plant population density than all other varieties at the mid-summer seeding date with no genotype differences on the other dates. Flowering measurements taken in July (2013 only) indicated that the two cultivars, Purple bounty and Purple prosperity, had close to 100% of plants producing flowers while all other varieties had almost no flowers. Therefore, Purple bounty and Purple prosperity appear to be the only cultivars tested that were vernalized under the conditions of this study. 4

6 Page 5 of 12 Biomass production measured at the end of the growing season was greatest for the early spring seeded vetch in both years (Table 1). Halde et al. (2015) determined that 7000 kg ha -1 of hairy vetch biomass was required to provide sufficient weed suppression for no-till organic seeding the following year. Only MB and MN seeded in early spring provided this amount of biomass in 2012 (Table 1). Biomass values were higher in 2013 due partly to more late season precipitation. Purple bounty had significantly less biomass than most other genotypes and Purple prosperity had less biomass than MB. End of season biomass for the early autumn plantings was not collected in 2012 since drought limited plant growth after establishment. Under more favourable late-season precipitation conditions in 2013, early autumn seeded hairy vetch plants still produced less than 200 kg biomass ha -1. It appears that early spring planting will maximize hairy vetch biomass production. Late-autumn weed biomass was measured for early spring and mid-summer (2013 only) plantings. Major weeds in both years were redroot pigweed (Amaranthus retroflexus), green foxtail (Setaria viridis) and yellow foxtail (Setaria glauca (L.) Beauv.). Where differences were observed, the lower biomass yielding Purple prosperity and Purple bounty had significantly more weed biomass than higher yielding MB and MN (Table 1). This demonstrates the value of genotype selection as a possible weed management strategy in hairy vetch cover crop production. Over-winter survival was observed only for the early-autumn seeded treatment since early spring and mid-summer plantings did not survive the winter regardless of genotype. For the early autumn planting, very few plants of varieties other than MN and MB survived (Table 1, 2013; visual observations only in spring, 2014, data not shown). Seed yields taken in 5

7 Page 6 of 12 August 23, 2014 from the surviving fall seeded plots showed that MN and MB produced 547 and 471 kg ha -1, respectively (data not shown). To our knowledge, hairy vetch cover crop seed yields have not been previously reported in Canada. If growers choose to seed MN or MB late in the season and do not want seed production, these hairy vetch plants will have to be terminated before seed set, which typically begins in late June. As expected, increasing seeding rate increased hairy vetch plant population density in both years (Table 2). Hairy vetch biomass was affected by seeding rate in all cases where it was measured; two dates in 2012 and one date in 2013 (Table 2). In 2012, the two highest seeding rates (90 and 120 seeds ha -1 ) had significantly more biomass than the two lowest rates (30 and 60 seeds ha -1 ) while in 2013, only the lowest seeding rate resulted in lower vetch biomass (Table 2). In both years, the highest seeding rates provided biomass levels close to 7000 kg ha -1, the minimum level for weed suppression (Halde et al. 2015). No differences between genotypes ( Oregon and Purple prosperity) were observed, nor were any genotype x seeding rate interactions recorded (Table 2). Across all treatments and sites, plant population density was positively correlated with hairy vetch biomass (r = 0.84; p = ). It was determined that greater than 50 hairy vetch plants m -2 were required for maximum hairy vetch biomass production (data not shown). Weed biomass was significantly reduced with increased seeding rate in both years (Table 2). The magnitude of the difference in weed biomass between low and high seeding rates was much greater in 2013 than It may be that the lower overall weed biomass in 2013 allowed hairy vetch at the higher seeding rates to be more competitive. Hairy vetch plant population and weed biomass were negatively correlated (r = -0.38; p = ) indicating that seeding rate is a possible weed management strategy in hairy vetch cover crop production. 6

8 Page 7 of 12 It was interesting to observe that hairy vetch biomass for the two highest seeding rates was similar (6489 kg ha -1 in 2012 and 6752 kg ha -1 in 2013) despite large differences in weed biomass; 4751 kg ha -1 in 2012 compared with 285 kg ha -1 in 2013 (Table 2). This observation suggests significant weed tolerance for hairy vetch, something that was previously observed (Halde et al. 2015). These authors observed significant hairy vetch growth in late summer and autumn, even under weedy conditions. They attributed this aggressive hairy vetch growth to it s viney nature, a trait that had been observed by Bedford (1900) over 100 years earlier, and the competitive advantage of the N-fixing ability of hairy vetch against non-fixing weeds. This preliminary study showed that 1) greater than 50 hairy vetch plants m -2 were required for maximum biomass yield, 2) delaying planting date to late summer increased overwinter survival and 3) growers should select genotypes of northern origin if seeking overwinter survival. Bedford, S. A Experimental Farm for Manitoba. Report of Superintendent. Annual Report of the Experimental Farms (Ottawa, Canada). Entz, M. H., Guilford, R. and Gulden, R Crop yield and soil nutrient status on 14 organic farms in the eastern portion of the northern great plains. Can. J. Plant Sci. 81: Halde, C., Bamford, K. C. and Entz, M. H Crop agronomic performance under a six-year continuous organic no-till system and other tilled and conventionally-managed systems in the northern Great Plains of Canada. Agric. Ecosys. and Environ. 213:

9 Page 8 of 12 Maul, J., Mirsky, S. B. Emche, S. and Devine, T Evaluating a germplasm collection of the cover crop hairy vetch for use in sustainable farming systems. Crop Sci. 51: Mills, G.F. and Haluschak, P Soils of the Carman Research Station. Special 10 Report Series No Manitoba Soil Survey Unit and Manitoba Land Resource 11 Unit. Agriculture Canada, Manitoba Department of Agriculture, and Department of 12 Soil Science of the University of Manitoba, Winnipeg, MB, Canada. Mirsky, S. B., Curran, W. S., Mortensen, D. A. Ryan, M. R.and Shumway, D. L Control of cereal rye with a roller/crimper as influenced by cover crop phenology. Agron. J 101: Mirsky, S. B., Ryan, M. R., Curran, W. S., Teasdale, J. R., Maul, J., Spargo, J. T., Moyer, J., Grantham, A. M., Weber, D., Way, T. R. and Camargo, G. G., Conservation tillage issues: Cover crop-based organic rotational no-till grain production in the mid-atlantic region, USA. Ren. Ag. and Food Sys. 27: Plant fact sheet: Hairy Vetch. USDA/NRCS (accessed U.S. National Plant Germplasm System (accessed April 25, 2016). 8

10 Page 9 of 12 Table 1. Effect of seeding date and genotype on hairy vetch establishment, flowering, biomass, and weed biomass production at Carman, MB in 2012/13 and Biomass 2012 Nov 16 Biomass 2013 Oct 9 Seeding date (D) Genotype (source) a Plant density 28 DAP b Flowering c 2012 Jul 17 Hairy vetch Weeds Plant density2013 May 15 Plant density 28 DAP Hairy vetch Weeds plants m kg ha plants m -2 plants m kg ha Mid-May Oregon b 4163b 7040ab ab 698 Wash b 4312b 6374b ab 493 PP a 3925b 8289a bc 757 PB a 1578c 8312a c 909 MB b 6697a 5823b a 180 MN b 7081a 5655b ab 462 avg b Pr > F genotype ns ns 9 Mid-summer Oregon b 60c 1328ab 2116 Wash b 69b 1560ab 1862 PP b 73b 1704a 1780 PB b 58c 1458ab 1788 MB b 88a 1217bc 1899 MN a 53c 950c 2061 avg a Pr > F genotype < ns Early autumn Oregon b Wash b PP b PB b MB a MN a avg Pr > F genotype ns ns - Pr > F D < Genotype <.0001 < <.0001 <.0001 <.0001 ns D*genoty pe <.0001 <

11 Page 10 of 12 Note: Means with a column not sharing a common lowercase italic letter differ significantly at the P<0.05 level. a Genotype (source): Oregon - Oregon grown; Wash. - Washington grown; PB - Purple Bounty; PP - Purple Prosperity; MB - Manitoba grown; MN - Minnesota grown. b Days after planting. c Flowering - 1 to 5 rating with 1=no flowers and 5=100% flowers. 10

12 Page 11 of 12 Table 2. Effect of seeding rate and genotype on hairy vetch establishment, biomass, and weed biomass production at Carman, MB in 2012 and Genotype (G) a Biomass, Aug 21 Biomass, Oct 16 Biomass, Oct 8 Seeding Plant Hairy Hairy Plant Hairy rate by density vetch Weeds vetch Weeds density vetch Weeds PLS m -2 plants m kg ha plants m kg ha PP avg. 48a 729a Oregon avg. 42b 579b d 360b 3415a 2490c d 4654b 888a 60 36c 528b 3075ab 4002b c 6073a 576b 90 53b 837a 2790b 6059a b 6327a 423bc a 891a 2871b 6489a a 6752a 285c avg Pr > F G ns ns ns ns ns ns Rate < <.0001 ns < <.001 G*Rate ns ns ns ns ns ns ns ns Note: Means with a column not sharing a common lowercase italic letter differ significantly at the P<0.05 level. a Genotype (source): Oregon - Oregon grown; PP - Purple Prosperity. b Seeding rate (pure live seed/m 2 ): Purple Prosperity 30 PLS = 10 kg ha -1 (2012), 9 kg ha -1 (2013); Oregon 30 PLS = 8 kg ha -1 (2012 and 2013). 11

13 Page 12 of 12 12